Predictions

Predictions are the core output of nirs4all. Every time you run a pipeline, nirs4all evaluates each model on each cross-validation fold and partition, storing the results as prediction records in the workspace. These records contain scalar scores for ranking, full arrays for visualization, and links to the trained model chain for export and replay.

The Prediction Lifecycle

Train               Store                Query              Export             Predict
nirs4all.run() ---> store.duckdb ------> result.top(n) ---> result.export() -> nirs4all.predict()
                    (predictions,         result.filter()    "model.n4a"        (new data)
                     chains,
                     artifacts)

  1. Train – nirs4all.run() executes your pipeline, fits models on each fold, and evaluates on train/val/test partitions.

  2. Store – Every prediction record (scores, arrays, chain reference) is persisted in the workspace database (store.duckdb).

  3. Query – Use result.top(n), result.filter(), result.best_rmse, etc. to find the best models.

  4. Export – Export the best model as a .n4a bundle for sharing or deployment.

  5. Predict – Apply the trained model to new data with nirs4all.predict().

Quick Start

import nirs4all
from sklearn.preprocessing import MinMaxScaler
from sklearn.cross_decomposition import PLSRegression

# 1. Train
result = nirs4all.run(
    pipeline=[MinMaxScaler(), PLSRegression(10)],
    dataset="sample_data/regression",
)

# 2. Check results
print(f"Best RMSE: {result.best_rmse:.4f}")
print(f"Best R2: {result.best_r2:.4f}")

# 3. Export best model
result.export("best_model.n4a")

# 4. Predict on new data
preds = nirs4all.predict(model="best_model.n4a", data=X_new)
print(preds.values)

Documentation Sections

Section

What You Will Learn

Understanding Predictions

What predictions are, how they are stored, key concepts (chains, partitions, scores)

Making Predictions

How to predict on new data from a RunResult, exported bundle, or chain ID

Analyzing Results

Querying, filtering, ranking, and visualizing prediction results

Exporting Models

Exporting models as bundles, scripts, or configs for sharing and deployment

Advanced Predictions

Transfer learning, retraining, SHAP explanations, batch prediction patterns

See Also